Stochastic maintenance optimization in additive manufacturing with digital twin integration

Additive Manufacturing (AM) plays a pivotal role in high-value production systems, but its extended build cycles make it particularly vulnerable to disruptions, resulting in substantial operational and economic consequences. To address these challenges, effective maintenance strategies are critical to ensuring operational reliability while minimizing costs. This paper proposes an integrated maintenance framework that combines stochastic degradation modeling, condition-based maintenance (CBM), and digital twin technology. The degradation process is modeled as a drift-diffusion (Wiener) process, and maintenance decisions are determined by adaptive inspection cadences and preventive maintenance thresholds, which are updated based on real-time health assessments of the system. The Remaining Useful Life (RUL) is estimated using a digital twin, which informs dynamic adjustments to both inspection intervals and maintenance triggers. Through an evaluation of four distinct maintenance strategies, the study demonstrates that adaptive policies (specifically those that integrate both adaptive cadence and adaptive threshold controls) result in significant reductions in lifecycle costs, minimized downtime, and enhanced operational reliability. The proposed framework offers a robust, data-driven approach for optimizing maintenance scheduling in AM systems. It provides actionable insights that reduce downtime, improve system reliability, and lower operational costs, making it an invaluable tool for industrial applications of AM.